SCRATCH is a web server designed to predict protein tertiary structure and structural features. The server includes various predictors for secondary structure, relative solvent accessibility, disordered regions, domains, disulfide bridges, single mutation stability, residue contacts, and tertiary structure. Users can submit an amino acid sequence and select desired predictions, which are then emailed to the user. The predictors use machine learning methods, evolutionary information, fragment libraries, and energy functions to make predictions. The server has been continuously updated with new methods and improvements, and it handles a significant number of jobs from around the world. The performance of the predictors varies, with some achieving high accuracy in specific categories. The server is particularly useful for targets without high sequence homology to solved structures. Future work includes developing a comprehensive tertiary structure predictor and improving current methods.SCRATCH is a web server designed to predict protein tertiary structure and structural features. The server includes various predictors for secondary structure, relative solvent accessibility, disordered regions, domains, disulfide bridges, single mutation stability, residue contacts, and tertiary structure. Users can submit an amino acid sequence and select desired predictions, which are then emailed to the user. The predictors use machine learning methods, evolutionary information, fragment libraries, and energy functions to make predictions. The server has been continuously updated with new methods and improvements, and it handles a significant number of jobs from around the world. The performance of the predictors varies, with some achieving high accuracy in specific categories. The server is particularly useful for targets without high sequence homology to solved structures. Future work includes developing a comprehensive tertiary structure predictor and improving current methods.